Effect of rider experience on thejumping kinematics of riding horses

Pippa NR Powers1,* and Anna M Kavanagh21Department of Physical Education and Sport Science, University of Limerick, Limerick,Ireland2Department of Life Sciences, University of Limerick, Limerick, Ireland*Corresponding author: pippa.powers@ul.ie

Submitted 20 June 2005: Accepted 14 October 2005 Research Paper

AbstractThe aim of this study was to examine the influence of an experienced rider and a novice rider on the stride kin-ematics of experienced riding horses. SVHS video recordings (50Hz) were made of ten experienced riding horsesjumping a 1.05m-high vertical fence. The horses were randomly assigned to jump the fence under two experi-mental conditions: ridden by an experienced rider and ridden by a novice rider. Three trials for each ridden con-dition were analysed, and the effects of the rider type on four kinematic variables were examined using a repeatedmeasures ANOVA. No significant differences were found between the riders for velocity and stride length duringthe approach, or for the take-off and landing distances from the fence. The results suggest that the riders bodyposition and body movement had no effect on the horses jumping kinematics as measured in this study, andthat each horse jumped the fence in its own manner, regardless of what the rider was doing. This is contraryto the current belief that a horses jumping technique is strongly influenced by the rider. These findings have rel-evance for both horses and riders, in that if an experienced horse does not respond to a riders instructions asexpected, then the implications for training of the horse and the rider are considerable.

Keywords: rider effects; horses; jumping; athletic performance

Introduction

Jumping is a popular equestrian activity that requires

significant communication between horse and rider.

Riders can communicate and interact with their

horses in a number of ways, mainly via changes in

the riders body position and use of the riders limbs.Other methods of interaction include the use of

verbal sounds and through physical contact using a

whip. When riders receive training and coaching, par-

ticularly early on in their careers, much emphasis is

placed on their body position and movement, as

these are considered to have a great deal of effect on

the way the horse moves.

Little research has been conducted on the effect ofthe rider on the kinematics of the horse. Biomechani-

cal analysis is a useful technique to quantitatively and

objectively measure any athletic performance. Some

work has examined the effects of a rider on the flat-

work kinematics of horses: for example, the effect of

a riders mass on the locomotion of horses during

treadmill exercise has been examined1, and the effects

of a riders mass on the ground reaction forces in trot-

ting horses has been analysed2. A more recent study

has examined the effects of a rider on the variability

of the horses gait while trotting on a treadmill3. In

equine jumping, some work has examined the effectsof a rider on the kinematics of jumping horses by

examining the differences between horses jumped

loose and ridden. It was found that the rider had a sig-

nificant effect on the horses body position at take-off,

over the fence and at landing4.

Fewer studies have examined the effect of riders of

different calibre on the kinematics of horses. The

differences between novice and advanced riders havebeen examined by analysing the head movement and

EMG activity of a number of trunk muscles5. This

study found that the novice rider had greater move-

ment of the upper body during the walk and trot.

No analysis, however, was conducted on the effects

of these riders on the horses movements. One

Equine and Comparative Exercise Physiology 2(4); 263267 DOI: 10.1079/ECP200568

qCAB International 2005

research group examined the movement patterns of

the rider and horse systems using riders of different

abilities6. They found that the professional rider

horse system had a more consistent motion pattern

than the recreational riderhorse system. Another

group examined the differences between a hobby

rider and a professional rider on the kinematics of a

trotting horse, and found that the horses ridden bythe professional rider had the highest trotting

speed, the longest stride length and the lowest pos-

ition of the head7.

To our knowledge, there is no evidence in the litera-

ture of any work examining the effects of riders of

different experiences on the jumping kinematics of

horses. Experienced riders have a tremendous ability

to control the locomotion and movement patterns oftheir horses. By lengthening or shortening the

horses stride, an experienced rider can position

a horse at an optimal distance from the fence at

take-off 8. By controlling the horses velocity during

the approach, the rider can help determine the

horses velocity at takeoff and, therefore, influence

the flight time over the fence and the distance

jumped9. It is generally accepted that riders of differ-ent calibre exist and that a good rider can achieve a

better performance from a horse than a poor rider10;

however, the effects of riders of different calibre on

the jumping kinematics of horses have not been

tested in the literature. The aim of this study was to

examine the effects of an experienced and a novice

rider on the linear kinematics during jumping in a

group of riding horses.

Methods

Ethical approval was obtained before the start of the

study from the University of Limerick Research

Ethics Committee.

RidersTwo riders were selected for the study. Rider 1 was

classified as an experienced rider (75 kg) with exten-

sive show jumping training and experience. This

rider had competed successfully at national level

show jumping for over 10 years. Rider 2 (63 kg)

was classified as a novice rider, and although compe-tent enough to jump safely, did not have any show

jumping experience. Visually, the riders were very

different in riding style and technique, with the experi-

enced rider being notably more balanced and more in

control of the horse than the novice rider.

HorsesTen horses were used for the study (age: 8 ^ 2 years;

height: 1.65 ^ 0.04m; weight: 498 ^ 39 kg). The

horses were selected on the basis of being similar in

size and experience. All horses were stabled in the

same yard and were used regularly in a large eques-

trian centre for daily riding lessons. The horses were

accustomed to jumping but were not used as compe-

tition horses. In accordance with the ethical approval,

owners/guardians of the horses consented to their

horses being used in the study.

Video recording protocolVideo recordings took place in a large, well-lit indoor

arena (90 30m). A halogen light positioned along-side the camera was also used to illuminate the record-

ing area. A single Panasonic AG450 camcorder

(Matsushita Electric Corporation of America, Secaucus,

NJ, USA) was set up perpendicular to the plane of

motion, c. 22m from the centre of a fence. The

zoom facility was used so that the field of viewwidth measured 10m, which allowed the recording

of the last approach stride and the jump stride. Two

reference cones were placed 4m apart on either side

of the fence. These were used to scale the coordinate

data in the recorded trials. This is a validated method

and has been used previously4.

A 1.05m-high vertical fence was used, and was

placed mid-way along one side of the arena. Threestandard show jumping poles were used: one was

used as a ground line, one as a diagonal filler and

one was placed horizontally at the 1.05m height.

This fence height was chosen as it was the maximum

height that the novice rider was comfortable with.

This fence height was also considered substantial for

all of the experimental horses to jump, since none of

them was a specialized jumping horse. The fencewas approached in a clockwise direction and from a

right-lead canter.

Videotaping took place over 2 days to avoid the like-

lihood of fatigue for the horses and the riders. Each

horse was randomly assigned to each rider and was

ridden once on each day (i.e. each rider rode five

horses on the first day and the remaining five on the

second day). Riders had a 30min interval betweenjumping sessions on each horse.

Each horse received a 15min warm-up consisting of

walking, trotting, cantering and a few practice jumps.

SVHS video recordings (50Hz) were made of the horse

and rider jumping the experimental fence. Riders were

requested to approach the fence at a canter and in a

manner they deemed most appropriate for a successful

jump. If the horse jumped the fence in an inappropri-ate manner, e.g. approached in the wrong lead leg or

spooked before the fence, then the trial was repeated

until three appropriate trials were collected.

Data analysisVideo recordings were analyzed using Peak Motus 3.2

(Peak Performance Technologies, Centennial, CO,

PNR Powers and AM Kavanagh264

USA). Four kinematics variables were examined, which

are illustrated and defined in Fig. 1. These variables (or

related variables) have been shown to be determinants

of success in previous studies examining the kin-

ematics of jumping horses11,12. They were selected

based on the relative ease of evaluating these variables

through qualitative analysis by a trained observer. The

distance measurements were calculated from the x-coordinate values of the respective points, e.g. foot

placements and fence base.

Descriptive statistics were calculated in Excel 2000

(Microsoft Corporation, Redmond, WA, USA). In

order to determine if the data from each horse

during each trial were normally distributed, measures

of skewness and kurtosis were calculated on the

horses data for each trial and rider type in SPSS(SPSS Inc., Chicago, IL, USA). Z values were deter-

mined by dividing the raw score for skewness or kur-

tosis by the appropriate standard error. Data were

considered to be normally distributed if the Z values

did not exceed ^2.013.

Inferential statistics were conducted using a general

linear model repeated-measures ANOVA in SPSS. With

a within-subjects design such as this, where the samegroup of subjects (i.e. the horses) is measured for the

same variables on a number of occasions, an ANOVA

with repeated measures is required to properly analyse

any differences in the data13. The statistical design

included two independent variables, namely, rider

(with two levels) and trial (with three levels) and four

dependent variables as defined in Fig. 1. A significance

level of P , 0.05 was set for the statistical tests.

Results

Each horse attempted the fence three to five times. From

the three trials selected as appropriate, all horses cleared

FIG. 2 Scatter plots illustrating the jumping attempts of eachhorse during each trial ( 3) for the experienced and novice riderfor each of the measured variables: a) SLAPPROACH; b) VxAPPR-OACH; c) LHFTO; d) TFFLAND

FIG. 1 Illustration and definition of variables measured in thestudy. VxAPPROACH Horizontal velocity of the last approach stride(measured from the time taken (i.e. number of frames) from theimpact of the trailing hind hoof in the last approach stride to theimpact of this trailing hind hoof at take off); SLAPPROACH Stridelength of the last approach stride (measured from the placementof the trailing hind hoof in the last approach stride to the contactof the trailing hind hoof at take-off); LHFTO Distance from theplacement of the leading hind hoof to the fence at take-off;TFFLAND Distance from the fence to the placement of the trailingfore hoof at landing etc.

Jumping kinematics of riding horses 265

the fence successfully during each trial for each rider.

The tests for skewness and kurtosis for each trial indi-

cated that the results of the horses for both riders were

normally distributed for each trial. For further infor-

mation, the scatter plots in Fig. 2 illustrate the results

of the variables for each horse during each rider trial.The descriptive statistics and the results of the ANOVA

are provided in Table 1. Although the horses ridden by

the experience rider had a slightly faster velocity and a

longer stride length during the approach, and took off

and landed further from the fence than the horses

ridden by the novice rider, no significant differences

were found between the riders for any of the measured

variables. The P value for trial indicates that there wereno learning or fatigue effects.

Discussion

This study set out to examine the effect of an experi-

enced rider and a novice rider on the stride kinematics

of a group of experienced riding horses. The approach

and take-off are critically important in determining the

outcomeof a jumpingeffort, and this has been documen-

tedmany times in the literature8,9. Interestingly, this pre-

sent study found no significant differences between tworiders of contrasting experience and skill for any of the

measured variables. The horses in this study approached

and jumped the fence in a similar manner for both riders

throughout all trials. This was a surprising result to the

authors, and does appear to contradict the general

acceptance among horse riders and trainers that the

rider input and influence are crucial for jumping.

There are a number of possible explanations forthese findings:

. The horses used in this study were experienced

riding school horses and not competition horses,

and perhaps just did not respond (as expected) to

the instruction or absence of instruction from

each rider. Considering that the kinematics of the

horses was consistent for each rider (ANOVA testfor trial effect was insignificant), it does, therefore,

seem likely that the horses jumped the 1.05m

experimental fence successfully and consistently

regardless of the riders movements or instructions

(this is visually evident from the scatter plots

in Fig. 2). This suggests that changes in the body

position of the rider do not have a significant

effect on some of the gross movements of the

horse during jumping. This is an interesting finding

considering that the riders body position and

movement are focused upon to a large extent

during jumping training of the rider. There is apossibility that the riding school horses may be

more accustomed to compensating for the unpre-

dictable movements of novice riders and ignore

the poor commands and signals given by these

riders.

. Because of the limitations of the novice rider, it was

not possible to test the difference between the

riders over a higher fence. Although this mighthave meant that some of the horses were jumping

submaximally and could perhaps tolerate any mis-

takes made by the novice rider, it does not explain

the fact that the horses approached the fence with

a similar speed and took off from a similar position

with both riders. It has been shown in previous

research that a rider can significantly affect the

jumping kinematics of young horses attemptingjust a 1m-high fence4, so there was no reason to

assume that the 1.05m fence height in this study

was too small to test this hypothesis. Previous

research has indicated that horses have a unique

and repeatable technique for jumping over a variety

of fence heights, even when ridden by experienced

riders14, so perhaps, as horses become more

experienced, this individual jumping techniquebecomes more permanent and less susceptible to

rider influence.

. The body mass of the experienced and novice

riders equated to c. 12 and 15% of the horses

body masses, respectively. It is not known what

effect a 3% difference in weight would have on

the jumping kinematics of the horse; however, for

this study, it was not deemed particularly importantas the stamina and endurance of the horses were

not considered as factors. In any case, evidence

in show jumping indicates that the weight of a

rider is relatively unimportant to performance,

since men and women compete on an equal basis

despite the differences in body size and strength.

. It is possible that the difference in rider experience

was not great enough to have an effect on thehorses, but this is unlikely considering the previous

Table 1 Descriptive statistics (mean ^ SD) and P values for the measured variables

Variable Unit Experienced Novice Rider P value Trial P value

VxAPPROACH m s21 6.08 ^ 0.53 5.80 ^ 0.63 0.271 0.926

SLAPPROACH m 2.56 ^ 0.53 2.49 ^ 0.38 0.612 0.837LHFTO m 1.52 ^ 0.38 1.49 ^ 0.31 0.772 0.846TFFLAND m 2.09 ^ 0.31 2.06 ^ 0.36 0.760 0.459

PNR Powers and AM Kavanagh266

histories of the riders and the notable visual differ-

ences in the riders styles and abilities. Neverthe-

less, these differences in rider style did not have

any effect on the kinematic variables measured

over a single fence. It may also be possible that

the riders had more of an influence on the horses

in the strides prior to the point where data collec-

tion began (i.e. preceding the final approach strideand jump stride), and this may have implications

for riders in negotiating a course or a combination

of fences rather than a single fence. This is an

aspect that could warrant further research.

. Finally, it may be that a single novice or experienced

rider is not aptly representative or prototypical of all

novice or experienced riders, and that the horses

may respond differently to a cohort of riders of vary-ing ability. Perhaps this is an area that also warrants

further research; however, previous work has

shown that a groupof horses (n 8) can respond dif-ferently to a single rider4, so there is no reason to

believe that the horses in this study could not have

responded differently to the riders involved.

Previous work has shown that the horses gait pattern isless variable with an experienced rider compared with a

novice rider3; however, this theorywas not supported in

this study. The relevance of these results may be con-

siderable for equine jumping research and training, but

at this stage it may be wise to err on the side of caution

and limit our conclusions to groups of riding school

horses, as other groups of horses (e.g. novice horses,

competition horses) may respond differently to thecommands and instructions of a rider. This is an aspect

that should be considered in future similar studies. How-

ever, if we accept the theory that as horses becomemore

experienced at jumping they become less influenced by

the riders instructions, then there are consequences for

the training of older horses and also for the training of

riders. If seasoned horses do not respond to the

riders instructions and simply jump in the mannerthey choose, then training regimes may have little ben-

efit for either the horse or the rider. Of course, for the

novice rider, an experienced horse is of great benefit.

However, for the more advanced riders wishing to

improve their riding ability and skills, a horse that does

not respond to their commands may not prove to be

very useful.

Acknowledgements

The authors wish to thank Clonshire Equestrian Centre

for providing the venue and the horses, and also Karen

Gardiner and Jennifer Dineen for assistance with the

data collection.

References

1 Sloet van Oldruitenbourg-Oosterbaan MM, Barneveld A andSchamhardt HC (1995). Effect of weight and riding onworkload and locomotion during treadmill exercise.Equine Veterinary Journal Supplement 23: 7175.

2 Clayton HM, Lanovaz JL, Schamhardt HC and Van Wessum R(1999). The effects of a riders mass on ground reactionforces and fetlock kinematics at the trot. Equine VeterinaryJournal Supplement 30: 218221.

3 Peham C, Licka T, Schobesberger H and Meschan E (2004).Influence of the rider on the variability of the equine gait.Human Movement Science 23: 663671.

4 Powers PNR and Harrison AJ (2002). Effects of the rider onthe linear kinematics of jumping horses. Sports Biomechan-ics 1: 135146.

5 Terada K (2000). Comparison of head movement and EMGactivity of muscles between advanced and novice horse-back riders at different gaits. Journal of Equine Science11: 8390.

6 Peham C, Licka T, Kapaun M and Scheidl M (2001). A newmethod to quantify harmony of the horserider system indressage. Sports Engineering 4: 95101.

7 Kapaun M, Peham C, Licke T and Scheidl M (1998). Theinfluence of the rider on the trotting motion of a riddenhorse. In: Riehle HJ & Vieten MM (eds), Proceedings ofthe XVI International Symposium on Biomechanicsin Sports. Konstanz: UVK-Universitatsverslag Konstanz.

8 Powers PNR and Harrison AJ (2000). How the rider canhelp show jumpers to have better performances? In: Lind-ner A (ed.), The Elite Show Jumper. Dortmund: LensingDruck, pp. 7990.

9 Clayton HM, Colborne GR, Lanovaz J and Burns TE (1996).Linear kinematics of water jumping in olympic show jum-pers. Pferdeheilkunde 12: 657660.

10 Klimke R (1989). Basic Training of the Young HorseLondon: JA Allen & Co.

11 Powers PNR and Harrison AJ (2000). A study of the tech-niques used by untrained horses during loose jumping.Journal of Equine Veterinary Science 20: 845850.

12 Colborne GR, Clayton HM and Lanovaz J (1995). Factorsthat influence vertical velocity during takeoff over a waterjump. Equine Veterinary Journal Supplement 18:138140.

13 Vincent WJ (1999). Statistics in Kinesiology. Champaign,IL: Human Kinetics.

14 Barrey E and Galloux P (1997). Analysis of the equine jump-ing technique by accelerometry. Equine Veterinary Jour-nal Supplement 23: 4549.

Jumping kinematics of riding horses 267